The visual system of primates is the most studied cortical sensory structure, yet it is still poorly understood. Future therapy for stroke victims or retinally blind patients will depend on our understanding of the basic biology of vision. One general principle that has been emerging from many studies is that different cortical areas process information differently. Specifically, cortical areas are arranged in a hierarchy, and higher areas appear to represent more complex stimuli. The present proposal represents a test of this idea. We wish to find out if one high-order region - the medial superior temporal area (MST) - is involved uniquely in a specific visual task, the analysis of complex optic flow patterns. We will test this idea with two physiological experiments on monkeys trained on such discriminations. First, we will manipulate signals in MST to see if performance on the task is affected. Secondly, we will record from single neurons in MST to formulate explicit, quantitative models of how the perception of motion arises from neuronal activity. The results of these experiments will allow a better understanding of information processing in higher areas of visual cortex.